High severity · published Mar 22, 2026
CVE-2026-4538
PyTorch pt2 Loading deserialization
7.8
CVSS
0.2%
EPSS
No
On KEV
Description
A vulnerability was identified in PyTorch 2.10.0. The affected element is an unknown function of the component pt2 Loading Handler. The manipulation leads to deserialization. The attack can only be performed from a local environment. The exploit is publicly available and might be used. The project was informed of the problem early through a pull request but has not reacted yet.
CVSS vector
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Attack vector
Local
Complexity
Low
Privileges
Low
User interaction
None
Confidentiality
High
Integrity
High
Availability
High
Scope
Unchanged
Exploitation
Is it being used?
Exploit prediction (EPSS)0.2% · higher than 16% of CVEs
Exploited in the wildNo evidence yet
CISA KEVNot listed
Ransomware campaign useNot known
Affected products
Which versions
PY
PyTorch ·
2.10.0 · source: vendor, NVD
References
NVD
https://nvd.nist.gov/vuln/detail/CVE-2026-4538
CVE.org
https://www.cve.org/CVERecord?id=CVE-2026-4538
Product
https://github.com/pytorch/pytorch/
Issue Tracking
https://github.com/pytorch/pytorch/pull/176791
Permissions Required
https://vuldb.com/?ctiid.352326
Third Party Advisory
https://vuldb.com/?id.352326
Third Party Advisory
https://vuldb.com/?submit.774681
More from PyTorch
7.5
CVE-2025-2148
PyTorch Tuple torch.ops.profiler._call_end_callbacks_on_jit_fut memory corruption
5.5
CVE-2025-3730
PyTorch LossCTC.cpp torch.nn.functional.ctc_loss denial of service
5.5
CVE-2025-3121
PyTorch torch.jit.jit_module_from_flatbuffer memory corruption
5.5
CVE-2025-2953
PyTorch torch.mkldnn_max_pool2d denial of service
5.3
CVE-2025-3001
PyTorch torch.lstm_cell memory corruption
